PROJECT SUMMARY: Human infants are faced with the challenge of learning about and engaging in their social world, including who belongs to their social in-group. Infants can infer social relationships by observing social interactions of others. For example, infants infer affiliation between novel individuals who speak the same language, synchronize movement, imitate each other, or comfort the same individual (Liberman, Woodward, & Kinzler, 2017; Powell & Spelke, 2018; Spokes & Spelke, 2017). However, little is known about how infants apply these inferences to interactions they see in their own social environment. For example, do infants pay attention to whom their caregivers affiliate with? Moreover, there remains several outstanding questions about infant's perceptions of social interactions. For example, infant's inferences about imitation are asymmetric?although 4 and 5-months-olds expect that an imitator will approach the individual they imitated, they do not expect an individual who was imitated to approach the person who imitated them. Likewise, although infants prefer those who imitate; they do not prefer those who are imitated (Powell, Spelke, 2018). This could be because infants only represent 'social actors'?the individuals who perform social actions?and not 'social targets'?the individuals whom social actions are directed toward. Or, it could be that infants prefer good candidates for affiliative partners, and that being imitated by a stranger is not informative to that end. After all, a person could be imitated without reciprocating or even noticing. In the current proposal three hypotheses are tested 1) Infants use their caregivers as 'references' when evaluating new people, preferring those who are affiliated with their caregivers 2) Infants can represent targets of affiliative social actions, but only prefer them if the infants know the social actor 3) Infant's get 'social value' from looking at or interacting with those who are affiliated with their caregivers. Preference will be measured through looking time and reaching. fNIRS (functional near-infrared spectroscopy) will be used to investigate whether infants' preferences reflect a feeling of 'social value' as opposed to curiosity or interest. Specifically, activation in the MPFC (medial prefrontal cortex) which is associated with social value in adults, will be compared to activation in the LPFC which is associated with processing novel information in adults. To sum, these studies will investigate a fundamental human social motivation?to find and affiliate with those in one's immediate social in-group. These studies will help us understand how children develop in-group biases, especially those found only in later childhood such as racial biases. Moreover, understanding the typical developmental trajectory of these social motivations will help us understand populations with atypical social development such as those with autism spectrum disorder.